britannia //finds.org.uk/documents/britannia/2015 (discussing 2014).pdf · britannia / volume 46 /...

Britanniahttp://journals.cambridge.org/BRI

Additional services for Britannia:

Email alerts: Click hereSubscriptions: Click hereCommercial reprints: Click hereTerms of use : Click here

II. Finds Reported under the Portable Antiquities Scheme

Sally Worrell and John Pearce

Britannia / Volume 46 / November 2015, pp 355 - 381DOI: 10.1017/S0068113X15000446, Published online: 08 September 2015

Link to this article: http://journals.cambridge.org/abstract_S0068113X15000446

How to cite this article:Sally Worrell and John Pearce (2015). II. Finds Reported under the Portable Antiquities Scheme.Britannia, 46, pp 355-381 doi:10.1017/S0068113X15000446

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/BRI, IP address: 87.74.139.172 on 17 Oct 2015

II. Finds Reported under the Portable Antiquities Scheme

By SALLY WORRELL and JOHN PEARCE

The Portable Antiquities Scheme was established in 1997 as an initiative to record archaeologicalobjects found by members of the general public and was extended to the whole of England andWales in 2003.1 Surveys of Roman period finds recorded by the PAS have been published inBritannia annually since 2004. This twelfth report gives an overview of the finds reported in2014 and of their character and distribution and publishes significant individual artefactsrecorded by Finds Liaison Officers in this year.

OVERVIEW

More than 50,000 metallic artefacts of Roman date were recorded on the PAS database in 2014. Asin previous annual statistics, this figure includes objects to which a date is attributed spanning thelate Iron Age to early Roman transition. Table 1 presents the numbers of artefacts of differentcategories recorded on the database by county, using older administrative boundaries forconsistency with previous reports; counties are grouped by PAS region. Table 1 divides theseartefacts into four principal groups, coins and brooches (the two numerically most significantcategories of finds), other personal ornaments and other non-ceramic finds.2 Not included inthe table are the additional 4,273 fragments of Roman pottery documented in 2014 as well asceramic and stone objects, including querns and small quantities of building material (tesserae,brick and tile, window-glass).

TABLE 1. NUMBERS OF NON-CERAMIC ARTEFACTS RECORDED BY THE PAS IN 2014 BY COUNTY ANDTYPE

Coins Brooches Other personaladornment

Other non-ceramicobjects

All non-ceramicobjects

WalesDenbigh 2 2 0 2 6Pembroke 0 2 0 0 2Wrexham 2 5 2 2 11Carmarthen 3 0 2 0 5Monmouth 9 3 2 3 17Powys 8 3 0 2 13Glamorgan 55 17 2 1 75Bridgend 3 0 0 0 3Gwynedd 0 1 0 0 1Newport 0 2 0 0 2Swansea 0 0 0 2 2Flintshire 1 0 0 0 1

Continued

1 S. Worrell, ‘Roman Britain in 2006. II. Finds reported under the Portable Antiquities Scheme’, Britannia 38 (2007),303.

2 This replaces the reporting scheme used between 2004 and 2011 in these summaries where data were presentedaccording to the major functional categories typically used for the publication of Roman period small finds. Readerswishing to explore regional variability among non-brooch finds in more detail are referred to these earlier summaries.

© The Author(s) 2015. Published by The Society for the Promotion of Roman Studies.

TABLE 1. CONTINUED

Coins Brooches Other personaladornment

Other non-ceramicobjects

All non-ceramicobjects

Conwy 1 2 0 0 3Rhondda CynonTaf

1 0 0 1

Anglesey 2 1 0 0 3EnglandNorthumbd 16 10 0 0 26Tyne & Wear 1 1 0 2Co. Durham 40 4 28 125 197N Yorks. 326 81 19 44 470E Yorks. 644 77 34 35 790S Yorks. 780 18 4 8 810W Yorks. 121 14 1 6 142N Lincs. 425 21 11 24 481NE Lincs. 9 2 0 1 12Cumbria 96 16 1 6 119Gt. Manchester 0 0 2 0 2Lancs. 35 6 1 8 50Merseyside 7 1 1 0 9Cheshire 94 35 9 26 164Notts. 124 29 8 7 168Derbys. 26 11 0 2 39Herefs. 10 12 1 3 26Shrops. 52 20 1 6 79Staffs. 107 41 2 13 163W Midlands 3 3 0 1 7Leics. 264 36 3 13 316Lincs. 1522 154 55 82 1813Rutland 170 6 6 3 185Worcs. 526 82 21 5 634Warwicks. 25 35 13 11 84Northants. 264 22 23 13 322Norfolk 2047 364 115 227 2753Suffolk 2737 229 48 71 3085Cambs. 2121 34 11 16 2182Essex 620 43 24 24 711Beds. 164 30 12 4 210Herts. 808 22 19 21 870Bucks. 479 31 9 29 548Oxon. 891 70 23 22 1006Gt. London 67 2 4 5 78Hants. 2490 75 16 24 2605Berks. 28 20 2 4 54IOW 347 26 4 10 387Surrey 285 17 7 5 314W Sussex 251 23 9 11 294E Sussex 69 17 4 2 92Kent 301 10 6 17 334Wilts. 5285 215 39 98 5637Glos. 930 118 18 19 1085Avon 61 25 6 5 97Somerset 483 57 10 6 556Dorset 507 105 18 31 661Devon 22274 6 3 8 22291Cornwall 89 20 0 0 109Total 49108 2333 660 1113 53214

SALLY WORRELL AND JOHN PEARCE356

As in previous years coins were by far the most common Roman artefact recorded in thedatabase in 2014, accounting for more than 92 per cent of the total metallic finds.3 The totalnumber of Roman coins documented by the scheme now stands at over 190,000, revealing thecontinuing impact of the initiative to record large assemblages of Roman coins in toto.4

The number of coins documented, the largest for any single year so far recorded, as well as thepercentage of finds that they account for, is the consequence of the documentation of severalvery substantial hoards on the database in 2014, comprising c. 30,000 of the Roman coinstabulated in Table 1. Eight individual hoards included more than 100 coins; of the fourcontaining more than 1,000, the most spectacular was the assemblage of 22–23,000 nummifrom Seaton Down, East Devon, one of the largest hoards of Constantinian nummi documentedacross the Empire.5 Otherwise the regional distribution of the coins reported for 2014 isgenerally similar to that of previous years, with significantly greater quantities being recordedin the eastern and some central and southern counties of England than in northern and westernEngland and Wales. As for 2013, large hoards explain the exceptions to this general pattern.6

The 2,333 brooches recorded for 2014 account for just over 4 per cent of all finds recorded inthis year, a lower than usual percentage explained by the inclusion of data from large coin hoardsin the coin totals reported. The total number of brooches now documented by the Scheme since2003 stands at well over 20,000.7 The distribution of brooch finds documented in 2014 alsoclosely resembles the pattern reported in previous years. Many more brooches are documentedin the counties of eastern England from North Yorkshire to Suffolk and in some centralsouthern and south-western counties, especially Gloucestershire and Wiltshire this year, thanelsewhere. The phenomenon observed in previous reports of significant regional variability inthe ratio of coins to brooches is again documented in 2014; brooches recurrently make up morethan 20 per cent of metal finds in the counties of the west and north-west Midlands.8 Otheritems of personal adornment (mainly finger-rings, bracelets and beads) account on average forc. 1 per cent of all non-ceramic finds. The percentage documented for individual countiesvaries markedly, but without the regional patterning observed for brooch finds. Leaving asidethe chance discovery of hoards, burials or related deposits (see below), the distribution of allfinds generally demonstrates a similar regional variability to previous years. This pattern is nowestablished on the basis of data collection over more than a decade and, as has been previouslynoted, is the product of many factors both historic and contemporary. These include thesocio-economic character of Roman period societies, past and current agricultural practice andthe varying intensity of metal-detector use across England and Wales, the relative significanceof which continues to be assessed.9

3 A selection of the most important coins is published annually in the British Numismatic Journal by S. Moorhead.4 S. Moorhead and P. Walton, ‘Coins recorded with the Portable Antiquities Scheme: a summary’, Britannia 42

(2011), 432–7; S. Worrell and J. Pearce, ‘Roman Britain in 2013. II. Finds reported under the Portable AntiquitiesScheme’, Britannia 45 (2014), 398.

5 These include: Seaton Down, Devon (PAS-D7EA4C/2013 T763); Ely, Cambs. (CAM-E70314); Riddlesden, WYorks. (SWYOR-810366); Forest of Dean (GLO-24A5E0); East Boldre, New Forest (HAMP-8C0F33); Hamering,Lincs. (DENO-6F8801/2014 T629); Ecclesfield, Sheffield (SWYOR-9E9660); Pewsey, Wilts. (BERK-637CB6).

6 Hoards continue to be published in full through the Coin Hoards of Roman Britain series, as well as the individualcoins being reported on the PAS database. A collaborative project on hoarding between the British Museum and theUniversity of Leicester continues; R. Bland, ‘Hoarding in Britain: an overview’, British Numismatic Journal 83 (2013),214–38; idem, ‘Hoarding in Iron Age and Roman Britain 2: the puzzle of the late Roman period’, British NumismaticJournal 84 (2014), 9–38.

7 Worrell and Pearce, op. cit. (note 4, 2014), 399.8 Worrell and Pearce, op. cit. (note 4, 2014), 399–400, with references to earlier reports.9 A Leverhulme-funded project on the distribution of PAS findspots and the factors determining them continues:

Portable Antiquities Scheme database as a tool for archaeological research. http://www.britishmuseum.org/research/research_projects/all_current_projects/pas_in_archaeological_research.aspx.

II. FINDS REPORTED UNDER THE PORTABLE ANTIQUITIES SCHEME 357

http://www.britishmuseum.org/research/research_projects/all_current_projects/pas_in_archaeological_research.aspx



In a small number of cases circumstances allowed for archaeological fieldwork associated withfindspots of significant discoveries, both coin hoards and burials, shedding further light on contextassociation and depositional process. Excavation following discovery suggested that the SeatonDown hoard (PAS-D7EA4C/2013 T763), comprising 22–23,000 copper-alloy coins, mostlyConstantinian nummi struck between A.D. 330 and 341, had been deposited inside a flexiblecontainer and was associated with three likely iron ingots of Roman date. At Riddlesden, WYorks. (SWYOR-810366), excavation and controlled detection at the site of a scatter of a smallnumber of denarii produced 110 further examples, dated from the reign of Vespasian to that ofMaximinus Thrax. The composition of the group, as well as its location, suggests that they mayperhaps be coins missed when the Bingley (Elam) Hoard was discovered in 1775 in the sameapproximate area. A hoard from Pewsey, Wilts. (BERK-637CB6), of c. 3,000 coins wascontained within a greyware jar which had been deposited in a small pit and covered with largelimestone blocks. The hoard was block-lifted during excavation for further analysis; the coinsso far identified are of fourth-century date. This year’s discoveries also include two significantburials found through metal-detecting and subsequently excavated. At Kelshall, Herts., threecopper-alloy vessels (a dish and two jugs) found by detecting were shown to have accompanieda cremation burial with a worn bronze coin (issued A.D. 174–5) in a hexagonal glass bottle.The burial was also furnished with a rich glass vessel assemblage, including two mosaic glassdishes, as well as a lamp and hobnailed shoes.10 A burial of similar date was also excavatedfollowing the discovery of copper-alloy vessels in detecting at Creslow, Bucks.11 Thefurnishings include ceramic, bronze and glass vessels, and an iron object, perhaps a lamp orlamp-holder, as well as the cremated remains placed with an intaglio and hobnails in a ceramicvessel. The handle of the bronze jug is richly decorated. Both the Kelshall and Creslow burialslie on the northern and western margins of the distribution in south-east England ofwell-furnished cremation burials of early to mid-Roman date with similar assemblages.12

While not found during controlled excavation, other objects from likely structured depositshave also been documented. The artefacts documented during continuing cataloguing of themany Roman finds from the river Tees at Piercebridge include examples of shoes withwell-preserved patterning in the hobnails embedded in the sole.13 A substantial deposit offourth-century ironwork was found at Marlborough, Wilts. (WILT-0E9BA9), including whatX-radiography has helped to reveal as c. 18 tools, including awls, a knife, a mower’s anvil,shears, lock elements and a linch pin. In its location and composition it is very closely relatedto the ironwork hoards in the southern group of late Roman date identified by Manning fromexcavated examples.14

ARTEFACT DESCRIPTIONS

The entries below set out some individual highlights of the past year’s discoveries recorded bymembers of the PAS and Treasure Department at the British Museum.15 These have beenselected because of their contribution to the understanding of object type, distribution or

10 K. Fitzpatrick-Matthews and J. Watters, ‘A grave at Kelshall: recreating the glory of a wealthy Roman citizen’,British Archaeology 142 (2015), 14–21.

11 M. Pitts, ‘Altar scene on Buckinghamshire Roman jug’, British Archaeology 142 (2015), 6.12 J. Pearce, Contextual Archaeology of Burial Practice: Case Studies from Roman Britain, BAR British Series 588

(2013), 111–29.13 e.g. BM-4AD3CB. For previous work on the Piercebridge assemblage see P. Walton, ‘The finds from the river’, in H.

E.M. Cool and D. Mason (eds), Roman Piercebridge. Excavations by D.W. Harding and Peter Scott 1969–81 (2008),286–93.

14 W.H. Manning, ‘Ironwork hoards in Iron Age and Roman Britain’, Britannia 3 (1972), 224–50.15 The object descriptions present revised versions by the authors of this report of the database entries made by PAS

staff and others. For almost all objects further discussion concerning their form and significance is also added here.


iconography, in some cases being items not previously recorded in the repertoire of small findsfrom the province. The reference number in brackets associated with each record is the PASunique identifier, which can be used to consult these and other object records on the PASwebsite: www.finds.org.uk. Four items were also treated as treasure cases and their Treasurenumber is also given in the format of year (20XX) plus reference number (TXX).16

The objects presented below belong to various categories, including anthropomorphic andzoomorphic figurines, personal ornament, household objects and tools. They are mostly copperalloy, but also include occasional examples in gold, silver and lead, and some carry enamelleddecoration. As in previous years they exhibit wide variety in iconography, style and technique.Their date extends from the transitional phase between the Iron Age and Roman periods to LateAntiquity. In detail they comprise the following: six anthropomorphic and zoomorphic figurines,including the first example from Britain of a Hellenistic genre piece, the ‘boy and goose’;personal ornament including bow brooches, zoomorphic and disc brooches, some with elaborateenamel inlay, as well as a ring with surviving intaglio; other artefacts including elements ofbronze vessels, a significant medical instrument find, a knife-handle with unusual enamelleddecoration, a zoomorphic châtelaine head, mounts and terrets from horse gear and furniture, alock of a type known only from PAS finds and a unique lead container for a small coin hoard.17

NORTHUMBERLAND

(1) Whittington (NCL-2B2752) (FIG. 1).18 A copper-alloy zoomorphic brooch of a horse of second- tothird-century A.D. date, 33.7 mm long, 26.1 mm high and 4.68 mm thick. The brooch is largelycomplete, missing only the back leg and the pin. The horse moves right with head lowered. On thehead visible features include an outsize and misplaced eye, a small slit mouth and, perhaps, anelement of harness above the eye. A full mane, its strands delineated, runs from the base of the neckto the top of the head where the forelock is braided into a topknot. At the shoulder two raisedlines run parallel from the mane to foreleg, perhaps rendering a harness element (a breastband?).The slender body ends in an expanded haunch and a short tail. The flank of the horse is recessed, asif for the attachment of some other element (e.g. a saddle and/or rider) but no trace of this isvisible. On the reverse a hooked catchplate and a lug with a perforation for a pin survive, with tracesof iron corrosion on either side of the lug. Traces of filing are also visible on the reverse.

FIG. 1. Whittington, zoomorphic brooch (No. 1). Scale 1:1. (Photo: R. Collins; © R. Collins and School of History,Classics and Archaeology, Newcastle University)

16 Throughout the year staff in the British Museum, in particular Ralph Jackson and Richard Hobbs, have providedinvaluable advice in the identification of individual objects. As ever we are especially grateful to Martin Henig forcomments on many objects. We also wish to record the help of Don Mackreth in identifying the Birdlip-type brooch.Janina Parol (British Museum) prepared images for publication. We thank too Barry Burnham for reading andcommenting on a draft text.

17 The geographical sequence in which objects are reported follows that set out in the ‘Roman Britain in 20XX. I. SitesExplored.’ section of Britannia.

18 Found by W. Marshall. Recorded by R. Collins and J. Pearce.


http://www.finds.org.uk

Unlike the majority of zoomorphic brooches in horse form, this example does not carryenamelled decoration. No direct parallel for this object could be found, although other horsebrooches also lacking enamel decoration similarly render the mane and topknot in detail.19

These other brooches lack a breastband of the type shown on the Whittington example (if thisis correctly identified as such); the same feature is, however, visible on some enamelledbrooches.20 It might be equivalent to the breastband of the type depicted, for instance, oncavalry tombstones, such as the monuments of the first-century A.D. equites Titus FlaviusBassus, Cologne, or Insus, Lancaster.21 Other well-presented mounts, for example as ridden bythe rider-god in the Stow-cum-Quy figurine (Cambs.), also have a flamboyant forelock andbreastband.22

DURHAM

(2) Piercebridge (NCL-9FFE11) (FIG. 2).23 An elaborate copper-alloy knife-handle, 85 mm long,with a maximum diameter of 16 mm. Much of the handle’s surface is decorated with a grid ofnarrow rectangular cells (three by eight). Enamel decoration survives in alternating cells; the othersare now empty, though in some there are traces of a pale green substance, perhaps an adhesive.The enamel-bearing cells in the middle block are filled with seven bands in alternating red andblue. The outer cells with enamel are filled with three thicker bands, again alternating between redand blue. There is no reserve metal between the different coloured enamels. At one end theknife-handle terminates in a circumferential rib followed by a flanged terminal which flattens atthe end. At the other, the handle extends to an hourglass moulding, beyond which extend two flat,rectangular, partially preserved plates with corrosion between them. One plate may have arivet-hole, although this is obscured by the poor preservation of the junction with the blade; theother carries two incised diagonal lines, perhaps traces of decoration.

A substantial number of knife-handles have been documented hitherto by the PAS. Many carryfigural decoration, but among these and among excavated examples geometric decoration is also

FIG. 2. Piercebridge, enamelled knife-handle (No. 2). Scale 1:1. (Photo: P. Walton; © P. Walton)

19 M. Feugère (ed.), Artefacts. Encyclopédie en ligne des petits objets archéologiques http://artefacts.mom.fr/fr/home.php, FIB-4282.

20 Feugère, op. cit. (note 19), FIB-4182; C. Johns, The Jewellery of Roman Britain: Celtic and Classical Traditions(1996), 177, fig. 7.16.

21 Bassus: CIL XIII. 8308; H. Galsterer, Die römischen Steininschriften aus Köln (2nd edn, 2010), no. 362, Ubi EratLupa 20697; Insus: RIB III.3185.

22 Worrell, op. cit. (note 1, 2007), 328–30, no. 26.23 Found by B. Middlemass and R. Mitchinson. Recorded by P. Walton.


http://artefacts.mom.fr/fr/home.php



common.24 However it has not been possible so far to find a specific parallel to this decoration ona knife-handle, although enamel decoration of this kind in grid or chequerboard format isparalleled among other objects, including vessels, brooches and harness-fittings.25

EAST RIDING OF YORKSHIRE

(3) Fangfoss (SWYOR-C74924) (FIG. 3).26 The head of a copper-alloy cosmetic set (châtelaine),dating from the first or second century A.D. It is 39.6 mm long, 34 mm wide and 9.7 mm thick. Theobject is hollow and modelled in the form of a chicken, with neck and tail extending from eitherend of a low hollow base. On either side of the head is an outsize eye formed by a circle of redenamel with a blue central dot. The head carries a worn crest above and terminates in two shortworn projections representing the beak and wattle. Two ridges at the base of the flattened tailseparate it from the body. The upper edge of the tail is decorated on both sides with rows ofsmall, closely-spaced punched crescents, being better preserved on the left side, where theirconfiguration in triangles is clearer. The function of the circular hole in the centre of the bodyis unknown. Four prominent ridges run around the circumference of the body, except for theflattened zone beneath the tail. From the base project two sub-rectangular lugs with perforationsin which remnants survive of the bar from which the toilet implements would have beensuspended.

Two other châtelaines have been documented by the PAS with some or all of the associatedcosmetic implements (i.e. nail-cleaner, ear-scoop and tweezers) still attached, one from Little

FIG. 3. Fangfoss, cosmetic set (No. 3). Scale 1:1. (Photo: A. Downes; © A. Downes)

24 Worrell and Pearce, op. cit. (note 4, 2014), 412–14, no. 14, Dodderhill, Worcs. for figural examples; the handle fromHursley, Hants. (HAMP-7A27F3) provides a more typical example of decoration dividing the handle surface into zones.

25 F. McIntosh, ‘A study into Romano-British enamelling’, The School of Historical Studies Postgraduate ForumE-Journal Edition 7 (2009), 1–18, https://www.societies.ncl.ac.uk/pgfnewcastle/6th-edition-of-the-historical-studies-postgraduate-forum-2010/; S. Worrell, ‘Enamelled vessels and related objects recorded to the Portable AntiquitiesScheme’, in D. Breeze (ed.), First Souvenirs: Enamelled Vessels from Hadrian’s Wall (2012), 71–84 (vessels); Worrelland Pearce, op. cit. (note 4, 2014), 410–11, no. 11 (a harness mount from Ludford, Lincs., DENO-FE87A5, with otherexamples); discussion of the Horton-cum-Peel brooch, Cheshire, (No. 8, below), includes references to other broocheswith similar decoration.

26 Found by P. Walsh. Identified and recorded by A. Cooper and J. Coulthard.


https://www.societies.ncl.ac.uk/pgfnewcastle/6th-edition-of-the-historical-studies-postgraduate-forum-2010/



Hadham, Cambs. (BH-1DB7F2), the other from Fincham, Norfolk (NMS-4FE992), both of the ‘barand shackle’ type.27 No close parallel to the Fangfoss artefact has been found among othersuspension elements. However, the existence of a suspension element whose form and decorationrecall those of enamelled zoomorphic brooches representing chickens, perhaps echoes thebetter-documented contemporary use of enamelled plate brooches for the same purpose.28

WAKEFIELD

(4) Thorpe Audlin (YORYM-D5BBD5) (FIG. 4).29 An incomplete copper-alloy bow Birdlipbrooch, with well-preserved enamelling. It is 57.9 mm long, 15.4 mm wide and 5.7 mm thick (itweighs 14.4 g). It has a near-circular head, below which is a raised collar, and a tapering bowwhich is poorly preserved towards its end where most of the enamel has been lost. Reservedmetal divides the head into cells laid out as a whirligig or similar shape, inlaid with colouredenamels; the cells are approximately comma-shaped and of varying size. They seem to havealternated between red and yellow, but on the top left most of the (putative) red enamel has beenlost and to bottom right the red cells join at the edge. The bow is also decorated with alternatingred and yellow enamel, organised in four pairs of rectangular cells which decrease in sizetowards the end of the brooch. Each pair of rectangular cells contains opposing semi-circularsettings in different coloured enamels, surrounded by reserved metal; each semi-circular settingcontains a central dot, also surrounded by reserved metal and in a different colour. The reverse isundecorated and retains the lug and a portion of the wire spring and catchplate. Although thebrooch type is well known, there are no known direct parallels to the decoration of this brooch.30

FIG. 4. Thorpe Audlin, Birdlip brooch (No. 4). Scale 1:1. (Photo: E. Cox; © E. Cox and York Museums Trust)

27 S. Worrell, ‘Roman Britain in 2003. II. Finds reported under the Portable Antiquities Scheme’, Britannia 35 (2004),26, no. 9; H. Eckardt and N. Crummy, Styling the Body in Late Iron Age and Roman Britain: a Contextual Approach toToilet Instruments (2008), 167–9.

28 Eckardt and Crummy, op. cit. (note 27), 170–2 for enamelled plate brooches; for zoomorphic chicken brooches seeM. Feugère, Les fibules en Gaule méridionale, de la conquête à la fin du Ve siècle ap. J.-C. (1985), 383–8, type 29c;C. Callahan, Cockerels in Romano-British Art, unpub. MA dissertation, University of Reading (2014), 34–9 (we thankHella Eckardt for drawing our attention to this dissertation).

29 Found by W. Colbeck. Identified and recorded by D. Mackreth, E. Cox and S. Worrell.30 D. Mackreth† (pers. comm.) suggests that the brooch belongs to the variant he identifies as Form 4.1b. D. Mackreth,

Brooches in Late Iron Age and Roman Britain (2011), I.12–13, II.8, pl. 5.


CHESHIRE

(5) Marbury cum Quoisely (LVPL-39BCF5) (FIG. 5).31 A flat handle of a copper-alloy pan,broken in two pieces but otherwise in good condition. The handle is 155 mm long and 43.53mm wide at the rounded terminal. From the latter the handle narrows slightly in the centrebefore widening and bifurcating where it grasps the rim of the bowl with two spreading arms.These arms take the form of elongated bird’s heads, separated from the rest of the handle by atriple ridge on both sides; an incised eye decorates each projecting arm before a beak-likeelement extends along the circumference of the now-missing pan. On the plain base of thehandle is a triangular projection which would also have attached to the body of the vessel. Atthe terminal end were originally four perforations (of which one has been lost) perhaps relatedto suspending the object.

The relief decoration on the handle is very worn, with a stylised plant element nearer theterminal, with three leaves or similar emerging from its base, and a figure advancing right,separated from the plant element by a groundline. A cloak or other object hangs from thefigure’s left arm; a curved object, perhaps a bow, rises from the right shoulder. Little can beidentified of the facial features. The figure appears to be winged and can therefore be identifiedas Cupid. This identification is supported by the much better-preserved relief images on twovery similar handles from Bonn and Neuss.32 Relief decoration documented on othercopper-alloy and silver handles of this type is otherwise varied, although the rendering of theprojecting arms which attach to the body of the vessel as elongated birds’ heads frequentlyrecurs.33

(6) Acton (LVPL-15E376; Treasure case 2014 T966) (FIG. 6).34 A conical lead-alloy container,containing a hoard of five silver denarii, issued between A.D. 194 and 198, sealed within by aplain lead disc forming a circular stopper. The container has a near-circular mouth, measuring45.2 mm by 40.2 mm (external dimensions); the cavity within it is 33.2 mm deep (internally)and the object weighs 50.5 g. The mouth has a wide, uneven rim, apparently formed by thelead alloy being folded or pushed back on itself. The cone tapers from the mouth to a roundedflattened terminal c. 5 mm in diameter, not quite centred. The terminal may have been cut offor broken; a perforation in the cone also indicates more recent damage. Among the variety ofcontainers used for coin hoards no direct parallel for an object of this kind is known.35

(7) Horton-cum-Peel (LVPL-F1F6CC) (FIG. 7).36 An almost complete copper-alloy disc broochwith well-preserved complex millefiori enamel decoration. It is 29.7 mm in diameter andweighs 7.8 g. Set within a raised border, the decoration is arranged within a somewhat irregularchequerboard grid in eleven rows and eleven columns. Within each there is a maximum ofeleven and a minimum of five squares. The individual squares are also in chequerboard form,with those based on larger and smaller squares set out alternately. The larger squares are laidout in a three by three grid within a red border and the smaller in a five by five grid within adark blue border. On the back is a double-lug loop from which a stub of the hinged pinprojects and a surviving catchplate.

Disc brooches of this type date to the later second and third centuries A.D. Several otherbrooches of this type, both whole and fragmentary, have now been documented, including PAS

31 Found by M. Tapp. Identified and recorded by V. Oakden and J. Pearce.32 H. Menzel, Die römischen Bronzen aus Deutschland III. Bonn (1986), 197–8, nos 541–2, Taf. 173.33 S. Tassinari, La vaisselle de bronze romaine et provinciale au Musée des Antiquités nationales (1975), 35–6, nos 28–

30, planches VIII–IX; J.M.C. Toynbee, Art in Britain under the Romans (1964), 320–1.34 Found by R. Moore. Identified and recorded by S. Moorhead and V. Oakden.35 Eleanor Ghey, pers. comm.36 Found by R. Cove. Identified and recorded by V. Oakden and J. Bayley.


FIG. 5. Marbury cum Quoisely, pan handle (No. 5). Scale 2:3. (Photo: V. Oakden; © V. Oakden and National Museums Liverpool)

SALLY

WORRELLAND

JOHN

PEARCE

364

examples from Standlake, Oxon. (BERK-9EC5C6), Claverdon, Warwicks. (WAW-D50D57), andWighill, N Yorks. (SWYOR-0F6696). An almost identical excavated example is reported fromAlcester.37 They are part of a wider group of dress and harness items as well as other artefactsbearing similar decoration.38

FIG. 6. Acton, lead container and denarii hoard (No. 6). Scale 1:1. (Photo: V. Oakden; © V. Oakden and NationalMuseums Liverpool)

FIG. 7. Horton-cum-Peel, disc brooch with millefiori enamel decoration (No. 7). Scale 1:1. (Photo: V. Oakden; ©V. Oakden and National Museums Liverpool)

37 J. Bayley and S. Butcher, Roman Brooches in Britain: A Technological and Typological Study Based on theRichborough Collection, Report of the Research Committee of the Society of Antiquaries of London 68 (2004), 130,illus. 31 and pl. 17.

38 See No. 3 above with references.


HEREFORDSHIRE

(8) Stretton Grandison (GLO-B172CF) (FIG. 8).39 A copper-alloy strap-mounted and skirtedterret of probable early Roman date. The terret is 87 mm long and 60 mm wide; it weighs134.9 g. The terret comprises two main elements, the ‘skirt’ and the near-circular loop. Theformer is perforated on both sides with a pair of holes. It rises at the edge into four upward-curving triangular projections, each ending in a spherical terminal. Beneath the ‘skirt’ is anintegral hoop. At the junction between ‘skirt’ and loop is a double row of stamped dots. Fromthis and on either side further double rows of dots extend to the tip of the skirt’s sideterminals. Above the loop is an integrally cast decorative element with two short angled armsforming a triangular projection. At the base of both arms is a flat round knop, at the apex is afurther spherical terminal. Rows of stamped dots run along the arms of the triangle and aroundthe collar at the base of the terminal.

This is a further example of an elaborate terret recorded by the PAS. Others include the examplewith horse-head terminals from Chollersbury, Bucks., and the double terret from Calbourne, Isle ofWight, with a croissant-like skirt.40

STAFFORDSHIRE

(9) Swindon (WMID-4CEB20) (FIG. 9).41 An incomplete copper-alloy Roman stud or mount, inthe form of a human head or mask, probably from a horse’s chamfron (face-guard). It is 20.7 mmlong, 22.0 mm wide and 10.2 mm deep (and weighs 6.4 g). A fringe of curls separates the headinto two elements, a head piece in the form of a high cap above and a face, perhaps of ayouth, below. The peaked head piece may be a ‘Phrygian’ cap of the type worn by divine orheroic figures, including Attis, Mithras and Ganymede. The features are simply rendered on arounded face, with large round circles for the eyes or irises, a triangular nose and a crudelymodelled mouth. The rear surface is uneven and corroded. At its centre an integral copper-alloyattachment spike is present, of which the end is now missing.

A mount recorded on the PAS database from Calbourne, Isle of Wight (IOW-60FC80), takes asimilar form, though it may be part of a larger object. A plaque carrying a very similar motif wasalso found during recent excavations of a deposit of Roman military metalwork at Caerleonlegionary fortress and is believed to be part of a cavalry chamfron (face-guard for a horse).42

The same motif also occurs on other military metalwork, especially parade armour, reinforcingthe possible connection of the Swindon and Calbourne finds to this milieu.43

LINCOLNSHIRE

(10) Kirton in Lindsey (LVPL-1F8252) (FIG. 10)44 A small copper-alloy figurine of Minerva,standing 59 mm tall. One arm is broken off just below the shoulder but the object is otherwisenear complete, though worn. The figure looks to her right and wears a high-crested helmet. The

39 Found by P. Gittings. Identified and recorded by K. Adams.40 S. Worrell, ‘Roman Britain in 2008. II. Finds reported under the Portable Antiquities Scheme’, Britannia 40 (2009),

301–3, no. 18; S. Worrell and J. Pearce, ‘Roman Britain in 2011. II. Finds reported under the Portable Antiquities Scheme’,Britannia 43 (2012), 381, no. 27.

41 Found by C. Pearson. Identified and recorded by T. Gilmore.42 P. Guest, ‘They came, they saw, they conquered. The Roman legionary fortress at Caerleon’, CAA History Magazine

9 (n.d.) 6, fig. 10b, http://www.aber.ac.uk/en/caa/web-projects/historymagazine9/. An apparently similar mount is reportedfrom Beaumes de Venise, Haute Provence, H. Rolland, Bronzes antiques de Haute Provence (1965), 151, no. 333.

43 J. Garbsch, Römische Paraderüstungen (1978), 83, R1-4, Taf. 41; S. Worrell and J. Pearce, ‘Roman Britain in 2010.II. Finds reported under the Portable Antiquities Scheme’, Britannia 42 (2011), 406.

44 Found by J. Davis. Identified and recorded by V. Oakden and S. Worrell.


http://www.aber.ac.uk/en/caa/web-projects/historymagazine9/



FIG. 8. Stretton Grandison, terret (No. 8). Scale 1:1. (Photo: K. Adams; © K. Adams and Bristol City Council)

FIG. 9. Swindon, stud or mount (No. 9). Scale 1:1. (Photo: T. Gilmore; © T. Gilmore and Birmingham City Council)


facial features are worn, though a linear mouth is visible. The left arm is held at an angle from thebody and the hand, bent backwards at the wrist, holds a spherical object. What survives of the rightarm suggests that it was extended at a right angle to the body, perhaps originally holding a spear.The goddess is clothed in a short-sleeved chiton which is secured with a belt, and also gatheredbelow the waist. To front and rear its folds reveal the movement of the legs beneath, the leftfoot forward, the right turned towards the right of the figure.

Minerva is among the commonest of female deities represented as a figurine to have beendocumented from the province, including examples recorded by the PAS.45 The pose of theKirton Lindsey figure, especially the marked twist to the right, is unusual however, anddifficult to parallel in figurines of Minerva documented in Britain and beyond.

OXFORDSHIRE

(11) Beckley (BERK-F1499B) (FIG. 11).46 A copper-alloy figurine of Mercury wearing a silvertorc and in muscular athletic form (78.24 mm high and weighing 64.8 g). The left arm hasbroken off above the elbow, but the figurine is otherwise in good condition. The thick-neckedgod stands with his weight on his right leg; his left is bent, in contrapposto pose. He is nakedapart from a cloak draped over his left shoulder; its folds are more clearly visible to the rear ofthe figure. In the palm of his left hand he holds a purse. The facial features are worn, butmodelled in some detail, including the rendering of the orbit (upper rim), lid and pupil for theeyes, long nose and full lips. The god has a luxuriant cap of hair, thicker at the fringes with itsstrands delineated. Two breaks on the top of the head indicate the position of the now missingwings. The silver torc around the god’s neck is a simple piece of round-sectioned silver wire,with flattened and pointed terminals decorated with hatching, perhaps recalling serpentiform

FIG. 10. Kirton in Lindsey, Minerva figurine (No. 10). Scale 1:1. (Photo: V. Oakden; © V. Oakden and NationalMuseums Liverpool)

45 E. Durham, ‘Depicting the gods. Metal figurines in Roman Britain’, Internet Archaeology 31 (2012), 3.29, http://dx.doi.org/10.11141/ia.31.2; Worrell, op. cit. (note 40, 2009), 301, no. 17, (WAW-C54295); Folkestone, Kent (KENT-178176).

46 Found by N. Mortensen. Identified by M. Henig. Recorded by A. Byard.


http://dx.doi.org/10.11141/ia.31.2



jewellery.47 The musculature is modelled in some detail, especially on the figure’s chest andabdomen, although the pectoral and trapezius muscles connect awkwardly to those of the upperarm. A fringe of pubic hair above the groin and nipples are also clearly rendered.

Mercury is the commonest occurring figurine from Roman Britain, represented by more than100 examples in Durham’s corpus and with numerous instances now documented by thePAS.48 In general the Beckley figure conforms to the common iconography for Mercury butexamples from the province more often show the god with petasos and cloak and the pursesuspended from the right hand. The Beckley figurine conforms to a type which is betterrepresented on the Continent, especially in northern and eastern Gaul, where the god is naked,with the occasional exception of a cloak, as in this instance, and holds the purse in the palm ofthe right hand rather than suspended from it.49

The torc is an occasional attribute of Roman deities. Separate torcs are documented inassociation with two other Mercury figurines from Britain, a silver example with thewell-known group from St Albans and a gold example from Richborough.50 They are morecommonly documented as an integral part of the figure. Examples include the ‘Fortuna’ figure

FIG. 11. Beckley, Mercury figurine (No. 11). Scale 1:1. (Photo: A. Byard; © A. Byard)

47 Johns, op. cit. (note 20), 44–7, 109–11.48 For Mercury figurines in Britain in general see Durham, op. cit. (note 45), 3.15. For examples reported to the PAS see

references in the discussion of a figurine from Nettleton, Lincs. (LIN-25CC02), S. Worrell and J. Pearce, ‘Roman Britain in2012. II. Finds reported under the Portable Antiquities Scheme’, Britannia 44 (2013), 55, no. 8.

49 S. Boucher, Recherches sur les bronzes figurés de Gaule pré-romaine et romaine (1976), 104–5, nos 159–68,planches 35–6; A. Kaufmann-Heinimann, Die römischen Bronzen der Schweiz. I. Augst, Forschungen in Augst 26(1977), 28–9, with an extensive list of occurrences; G. Faider-Feytmans, Les Bronzes romains de Belgique (1979), 60–4, nos 32–42, planches 14–12.

50 Boucher, op. cit. (note 49), no. 370, documents a further example from Mathay, Doubs; Durham, op. cit. (note 45),nos 55 (St Albans) and 284 (Richborough).


from Kettlebaston, Suffolk (SF-302EF1), a ‘mother goddess’ from Henley Wood and a male deity (?)from Vevey, canton Vaud.51 The torc may complement a votive offering through embellishment ofthe figurine with a more costly material; it may also be an attribute inherited from a pre-Roman deitywith which Mercury is assimilated in this case.52

(12) Wallingford (SUR-AC2EFB) (FIG. 12).53 An enamelled zoomorphic brooch in the form of arunning stag, 29.76 mm long and weighing 4.58 g. The animal’s long head is well modelled, withopen mouth and large oval eye framing a central dot. A small ear is incised below the antlers whichcurve together to form a crown-like loop. The neck is disproportionately large and the bodyelongated, expanding at the shoulder and rump; the legs are mostly missing and a triangularprojection indicates the creature’s sex. The back of the curving neck is notched and theincisions extend to the back and rear haunch but the principal decoration comprises twoenamelled panels. The larger extends from the head to the forelimb and along the body, thesmaller is circular and fills the rump. Within the larger panel, above the forelimbs, is a reservedcircular setting. On the flat back the catchplate largely survives along with the hinge, the lattercontaining the remains of an iron pin.

Similar examples are documented from pre-War excavations in Wroxeter and widely acrosscontinental Europe; this brooch may be of Continental workmanship and of second- tothird-century A.D. date.54 Other stag brooches of varying form are documented among PASfinds, though without close parallels to this.55 Similar representations of stags also occur inother media, for instance figurines; the example reported from Kirkby la Thorpe, Lincs.(LIN-E63C92) is not atypical of small-scale pieces.56

HERTFORDSHIRE

(13) Sarratt (BH-98132A) (FIG. 13).57 A largely complete copper-alloy figurine in the form of astanding bull, 58.4 mm long, 47.7 mm high, 18.3 mm wide and weighing 130.6 g. The solid-castobject is of realistic appearance, with well-modelled physique, but with little surviving surface

FIG. 12. Wallingford, zoomorphic brooch (No. 12). Scale 1:1. (Photo:D. Williams; © D. Williams)

51 E. Durham, ‘Style and substance: some metal figurines from south-west Britain’, Britannia 45 (2014), 45–6;A. Leibundgut, Die römischen Bronzen der Schweiz. III. Westschweiz, Bern und Wallis (1980), 45–6, no. 41, Taf. 53;S. Worrell, ‘Roman Britain in 2009 II. Finds reported under the Portable Antiquities Scheme’, Britannia 41 (2010),430–1, no. 15.

52 Boucher, op. cit. (note 49), 174–5, planches 7–9, including the famous cross-legged figure from Bouray.53 Found by D. Pearce. Identified and recorded by D. Williams and S. Worrell.54 D. Atkinson, Report on Excavations at Wroxeter (the Roman City of Viroconium) in the County of Salop, 1923–1927

(1942), 208, H26, fig. 36; Feugère, op. cit. (note 28), 385, 405, fig. 59, Type 29, no. 12b.55 Worrell and Pearce, op. cit. (note 48, 2013), 361–2, no. 15.56 Worrell and Pearce, op. cit. (note 48, 2013), 356, no. 9; Durham, op. cit. (note 45), 3.35.12.57 Found by G. Edmund. Identified by M. Henig and J. Watters. Recorded by J. Watters.


detail. Wear to the stubby head of the animal means that facial features are scarcely visible, theclearest surviving element being the eyes. The left horn is a little better presented than theright; between the horns are traces, perhaps, of tufts of hair. Below are regular oblique lineswhich may be further curls or a fabric band around the forehead. Beneath the lower jaw andneck sags a prominent dewlap; on the underside of the animal the penis is clearly visible. Themass of muscle at the shoulders and thick neck lend the beast a powerful appearance; the rumpis similarly muscled. Shoulders and rump taper into straight legs, though only one of these (leftrear leg) is complete. A tail hangs between the back legs, terminating a short distance from thesurviving hoof.

Bull/bovine figurines occur sporadically among the animal figures documented as late Iron Ageand Roman period small bronzes in Britain.58 Other standing bull or bovine figures documented bythe PAS include examples in soapstone from Atcham, Shrops., and in copper alloy fromWeaverham, Cheshire (LVPL-CB6114), Watlington, Oxon. (BUC-668F82), and Holbrook,Suffolk (SF-DCB627), in all three cases with three horns.59 Small hollow copper-alloy figuresof recumbent animals have been recorded from Marple, Stockport (LVPL-904BD3), and fromCalbourne, Isle of Wight (IOW-2CA926), the latter with a dorsuale.60 In 2014 a fragment of awhite marble statue of a cow, more than 370 mm long and lacking its head, perhaps of first- orsecond-century date, was discovered during drainage works in Lincoln’s northern suburbs. Thismay be a Grand Tour souvenir rather than an ancient import (LIN-2B67D5).

WEST BERKSHIRE

(14) Winterbourne (BERK-B60E47) (FIG. 14).61 A near-complete but worn copper-alloy figurineof a chubby, naked, winged boy with a large bird, perhaps a goose. The statue, weighing 167 g, is59.9 mm high and the disc that serves as the base is 35.6 mm wide. The boy, head inclined slightlyto the right, holds the goose’s neck with his right hand and places his left across its shoulders. The

FIG. 13. Sarratt, bull figurine (No. 13). Scale 1:1. (Photo: J. Watters; © J. Watters)

58 Durham, op. cit. (note 45), 3.35.2. Continental examples of bull figurines are referenced in the PAS record.59 Worrell, op. cit. (note 51, 2010), 420–1, no. 6.60 S. Worrell and J. Pearce, ‘Roman Britain in 2010 II. Finds reported under the Portable Antiquities Scheme’,

Britannia 42 (2011), 429–30, no. 24.61 Found by A. Ashford. Identified and recorded by A. Byard, M. Henig and J. Pearce.


boy’s left leg bends at the knee with the lower leg beneath the upper; the right leg extends forward,bending a little at the knee. The goose, with its plumage schematically rendered, appears to standon the boy’s left leg, its long neck reaching up towards him with its beak beneath his chin. Themodelling of the pair is relatively crude, the boy being a squat figure with a broad back fromwhich two small wings protrude below the shoulders. Shoulders and limbs aredisproportionately large. The head is much worn, but the figure appears to wear a cap beneathwhich a fringe of curls emerges on all sides. Of the facial features, almond-shaped eyes, broadnose, and small mouth are visible. The action and mood are difficult to interpret; the boy maybe struggling with the goose, trying to wring its neck perhaps, but the gesture may be moreaffectionate.

This is one of many Roman period copies of a well-known Hellenistic genre figure linked to the‘Boy with Goose’ sculpture attributed by Pliny (N.H. 34.84) to Boethos of Chalcedon and ofthird-century B.C. date, though ancient textual sources on the identity of the sculptor are noteasily reconciled. The subject group may have a longer history and the relationships betweenchild and animal vary from the kindly to the murderous. Other versions of the pair take bothmonumental and smaller-scale forms, the latter including examples in silver, terracotta andbronze.62 Like the Winterbourne instance, some other pairings assimilate the child with thefigure of Cupid through the presence of wings. Another miniature example of the samesculpture from Britain, not previously recognised as deriving from this Hellenistic genre group,also presents the child in the form of Cupid. This is one of several figurines from the Lexdentumulus, Colchester, constructed in the last decades of the first century B.C.63 The latter wereimported, but in the case of the Winterbourne figure the style may suggest that it was made inBritain. Whatever its purpose, whether a piece for demonstrating cultivation and connoisseurship

FIG. 14. Winterbourne, ‘boy with goose’ figurine (No. 14). Scale 1:1. (Photo: A. Byard; © A. Byard)

62 E. Gardner, ‘A statuette representing a boy and goose’, Journal of Hellenic Studies 6 (1885), 1–15; B. Ridgway, ‘Theboy strangling the goose: genre figure or mythological symbol’, American Journal of Archaeology 110 (2006), 643–8; J.J.Pollitt, Art in the Hellenistic Age (1986), 128–9, nos 132–3; R.R.R. Smith, Hellenistic Sculpture (1991), 136–40. The PASrecord also lists further examples.

63 Durham, op. cit. (note 45), no. 68, identified as example of Cupid; J. Foster, The Lexden Tumulus: a Re-appraisal ofan Iron Age Burial from Colchester, Essex, BAR British Series 156 (1986), 52–4, fig. 20.1, pl. 8.


or an image from a household shrine, this is another significant example of an image originating inHellenistic genre sculpture now documented by the PAS.64

MILTON KEYNES

(15) Moulsoe (WMID-0503C1) (FIG. 15).65 The base of a copper-alloy jug handle, dating to thefirst to second centuries A.D. The fragment is 49.9 mm long, 32.0 mm wide and 22.1 mm thick.The fragment comprises mainly the figured terminal where the handle is attached to the bodyof the vessel, as well as a short length of the ridged handle. The front of the terminal is cast inthe form of the head of a male youth, whose eyes look outwards and downwards. The hair iscurly and thick, rising in exuberant locks from the temples with its strands differentiated. Theivy leaves visible in the hair on the left and right sides of the head are part of a likely wreath.The face is full and rounded, with silver-inlaid eyes (and reserved circular depressions for thepupils) and a nose and lips in low relief, made less prominent through wear.

Generally similar faces and masks have been documented as handle terminals among other PASfinds, for example at Brockley, Suffolk (SF-452BA2), Codford, Wilts. (WILT-7E0308), andChislet, Kent (KENT-6E5FE6), as well as on older finds of complete vessels from Canterburyand Stanfordbury, Beds.66 The masks used as vessel escutcheons for the pans associated withthe workshop of the Cipii are more similar in spirit to that at the base of the Moulsoe handle.67

In this case the ivy leaves indicate the specifically Bacchic connotations of the object.

FIG. 15. Moulsoe, jug handle (No. 15). Scale 1:1. (Photo: T. Gilmore; © T. Gilmore and Birmingham City Council)

64 E. Bartman, Ancient Sculptural Copies in Miniature (1992), 41–2, 47; Boucher, op. cit. (note 49), 181–204,esp. 195–6, notes a widespread occurrence in Gaul, including group subjects. Other PAS examples include a figurine ofa sleeping boy from Saffron Walden, Essex (ESS-6F60D3) (Worrell, op. cit. (note 1, 2007), 331, no. 26) and a pendantor steelyard weight rendered as a grotesque head, Osbournby, Lincs. (LIN-1213A7) (Worrell and Pearce, op. cit. (note40, 2012), 365–6, no. 8).

65 Found by K. Hewitt. Identified by T. Gilmore, S. Worrell and J. Pearce.66 S. Worrell, ‘Roman Britain in 2005. II. Finds reported under the Portable Antiquities Scheme’, Britannia 37 (2006),

451, no. 19; J.M.C. Toynbee, Art in Roman Britain (1962), 175, nos 114 and 116, pls 130, 132; Toynbee, op. cit. (note 33),322–6, for proliferation of faces in this setting.

67 Kaufmann-Heinimann, op. cit. (note 49), 21–2, Abb. 3.


NORFOLK

(16) Woodbastwick (NMS-800B35) (FIG. 16).68 An unusual copper-alloy mount in the form ofthe bust of a youthful male. It is c. 80 mm tall and weighs 142.3 g. The youth wears a wreathof ivy leaves, with a gap at the centre and a winged cap, a petasos. The latter perches atop aluxuriant cap of hair, the thick curls of which frame the face. The facial features of theclean-shaven youth are well modelled. The ties of the wreath descend behind the ears to hangdown as far as the folds of the tunic-like garment the figure wears. The eyes are outsize andslightly asymmetrical; their eyelids, irises and pupils are rendered in some detail. The nose isbroad and the lips are full, while the ears are set a little too low on either side of the head. Theflat-backed object is hollow, indicating that this was probably a furniture mount; the small holeon the top of the cap may relate to the same function.

This is one of many recent figural finds from Roman Norfolk. The sensitive treatment of thissubject, together with the emphasis on the pupils, may suggest a Hadrianic or Antonine date for themount.69 The youth bears traits associated with Mercury and Bacchus, the petasos for the former,ivy leaves for the latter. Mounts documented from furniture and other objects occur in multiplefigural forms. These are frequently divine and Bacchic in character, although generic youths

FIG. 16. Woodbastwick, mount with hybrid figure (No. 16). Scale 1:1. (Photo: A. Marsden; © Norfolk CountyCouncil)

68 Found by G. Cook. Identified by A. Marsden.69 M. Henig, pers. comm.; A. Marsden, ‘Satyrs, leopards, riders and ravens’, in S. Ashley and A. Marsden (eds),

Landscapes and Artefacts. Studies in East Anglian Archaeology Presented to Andrew Rogerson (2014), 61, no. 41.


also occur.70 Bacchic examples from Britain include objects from Littlecote, Berks., TarrantHinton, Dorset, and Heversham, Cumbria (LANCUM-1747E5).71 No specific parallel to thiscombination of attributes of both deities on an object of this type could be found.

ESSEX

(17) Roxwell (ESS-33D3A2) (FIG. 17).72 An incomplete copper-alloy figurine, probably of a maledeity. It is 64.9 mm high, 48.1 mm wide (maximum extent) and 17.3 mm thick. It weighs 65.8 g.The figure stands with arms extended to the sides and bent at the elbows, the left upwards, andthe right downwards, both broken just after the elbow. The legs are also broken below the kneebut enough survives to suggest that the right leg was straight and the left leg bent. The figurewears a hooded cloak which ends in a peak or crest. The cloak is seemingly fastened at theright shoulder and hangs stiffly down the figure’s back (with a hole pierced in it at waistheight on the figure’s left side). Its folds are schematically rendered by a near parallel ‘V’-likedouble moulding from the shoulders to the base. A zig-zag runs along both long edges,indicating a decorative pattern or fine edging. On the figure’s right shoulder is a short quiver,at the top end of which arrow feathers are visible. Beneath the cloak the figure wears apleated tunic bound at the waist with a substantial belt. The disproportionately large headand its facial features are simply modelled; the face may be fringed with curls (or the liningof the hood).

No direct parallel has been found, although in its form and attributes this object echoes therepresentations of several other deities, personifications and other figures. While the wearing ofa hooded cloak is shared with the Genius Cucullatus, for example, and the ‘countryman’ fromTrier, the cloak worn by the Roxwell figure is different and in other respects he differs

FIG. 17. Roxwell, hunter figurine (No. 17). Scale 1:1. (Photo: K. Marsden; © Colchester and Ipswich MuseumService)

70 B. Barr-Sharrar, The Hellenistic and Early Roman Decorative Bust (1987).71 B. Walters and M. Henig, ‘Two busts from Littlecote’, Britannia 29 (1988), 407–10.72 Found by D. Allen. Identified and recorded by K. Marsden and J. Pearce.


from them.73 In its arms and clothing the Roxwell figure recalls the ‘hunter god’ identified fromsculptural representations in London and the Cotswolds, perhaps a syncretised version of Apolloand an indigenous god.74 However, unlike the ‘hunter god’, the Roxwell figure does not wear aPhrygian cap and the arrangement of the arms shows that he cannot have been drawing anarrow from a bow. Instead the cloak and pose recall some personifications of winter, bestknown in Britain from the seasons mosaic at Chedworth villa, Glos., in which the hoodedfigure holds a hare in its right hand and a branch raised in its left.75 The Roxwell figure may,therefore, be a variant representation of one of these divine beings, or alternatively render alocal god by drawing on the traits of multiple deities. Other examples of probable divinefigures documented by the PAS with attributes from varying sources include ‘Fortuna’ fromKettlebaston, Suffolk (SF-302EF1), and a ‘Minerva-Fortuna’ hybrid from Hinxworth, Herts.(BH-ED9F44).76 Alternatively the Roxwell figure may be a genre representation of a huntsman,of which other examples are occasionally recorded.77

KENT

(18) Elmsted (KENT-FE87D8) (FIG. 18).78 A copper-alloy tripod mount in the form of the bust ofa young male figure, his head turned slightly to the right. The mount is 78.7 mm long, 37.4 mmwide, 28.5 mm thick and weighs 134.8 g. The head has finely modelled hair and facial features,supported on a thick neck. At the crown of the head is a thick cap of hair, enclosed by tworows of thickly curling locks. Strands of hair fall between ear and face and the forelock curlshigh above the forehead. The facial features are well-modelled but worn. The bust, flat at theback, rises from a calyx, separated from the hollow rectangular base by three transversemouldings. There is a large irregular hole on the back, created by the breaking away of a likelyintegral hook which is no longer attached. The hole reveals that the mount is hollow; ironstaining can be seen within where the mount connected to the leg of the iron tripod to which itwould have been attached.

Other tripod mounts have been documented by the PAS across England. Examples fromPocklington, E Yorks. (YORYM-EC06D2), and Greetwell, Lincs. (LIN-1632D1), representyouths with curly heads of hair who are more easily identified as divinities. Other examples arereported from Rodmell, Sussex (PUBLIC-F077EF, a male bust) and Pickhill with Roxby, NYorks. (LVPL-CB8B04, a female bust).79 Further examples similar to the Elmsted bust arereported from Trier and Bois-et-Borsu (Liège, Belgium); the latter example is associated with avery well-preserved tripod from a rich second-century A.D. burial assemblage.80

73 Worrell and Pearce, op. cit. (note 43, 2011), 416–17, no. 12, describe an example of a Genius Cucullatus fromBuntingford, Herts. (BH-712AC1) and similar figurines. For the countryman figurine see H. Menzel, Die römischenBronzen aus Deutschland II.Trier (1966), 41, no. 86, Taf. 40.

74 R. Merrifield, ‘The London Hunter-God and his significance’, in J. Bird, M. Hassall and H. Sheldon (eds),Interpreting Roman London (1996), 105–13; P. Coombe, F. Grew, K. Hayward and M. Henig, CSIR Great Britain.1.10, Roman Sculpture from London and the South-East (2015), 42–4, nos 73–5, pl. 41; M. Henig, CSIR Great Britain.1.7, Roman Sculpture from the Cotswold Region, with Devon and Cornwall (1994), 37–9, nos 110–14, pl. 29.

75 S. Cosh and D. Neal, Roman Mosaics of Britain. Vol. IV Western Britain (2010), 54–8; an antiquities auction websiteshows an unprovenanced figurine of winter in similar form: https://www.liveauctioneers.com/item/36283119_rare-roman-bronze-personification-of-winter

76 Worrell, op. cit. (note 51, 2010), 430–1, no. 15; G. Burleigh and R. Jackson, ‘An unusual Minerva-Fortuna figurinefrom Hinxworth, Hertfordshire’, Antiquaries Journal 89 (2009), 63–7.

77 Durham, op. cit. (note 45), no. 1129, huntsman with deer slung over his shoulders, London.78 Found by T. Mills. Identified by R. Jackson and S. Worrell. Recorded by J. Jackson.79 Worrell, op. cit. (note 40, 2009), 288–90, no. 6, provides references to the handful of examples discovered during

excavation.80 Menzel, op. cit. (note 73), no. 286, Taf. 92; Faider-Feytmans, op. cit. (note 49), 185, no. 380, pl. 160.


https://www.liveauctioneers.com/item/36283119_rare-roman-bronze-personification-of-winter







(19) Eynsford (LON-B48905) (FIG. 19).81 A small copper-alloy figurine of Mercury, reclining ona flat base. The figurine is 52.8 mm long, 37.4 mm tall and 11.4 mm thick; it weighs 54.9 g. Thegod wears a winged traveller’s hat (petasos), holds a caduceus in his left hand and a purse in hisright. He wears a cloak over his left shoulder which extends down the left side of the rear of thefigure to the feet. A fringe of curly hair emerges from beneath the cap at front and back. The facialfeatures are crudely modelled, comprising two irregular depressions for eyes, a somewhat bulbousnose and an asymmetrical slit for the mouth. The god’s anatomy is also summarily rendered; littlemuscular detail can be seen on the torso, part of which is obscured by the outsize and claw-likeright hand and purse. Of the left arm only the hand is visible, grasping the caduceus halfway alongits shaft. The right leg covers most of the lower left. The metal on the base is slightly roughened,perhaps suggesting the figurine may have been attached to, or mounted on, another object.

Representations of Mercury are the commonest of all deities from the province. The god isusually depicted standing; seated or reclining figures such as this are rare, but other examplesoccur in similar forms, including some figurines from London as well as from continentalEurope, although these hold a libation bowl rather than a purse in their right hand.82

FIG. 18. Elmsted, tripod mount (No. 18). Scale 1:1. (Photo: J. Jackson; © Kent County Council)

81 Found by B. Robson. Identified and recorded by K. Sumnall.


HAMPSHIRE

(20) Tangley (HAMP-EC91E2; 2014T12) (FIG. 20).83 A late Roman gold finger-ring with goldfiligree and nicolo intaglio (an onyx with a blue surface on a dark ground). The ring measures22.3 mm by 21.6 mm and weighs 5.6 g. The bezel and hoop have broken on one side but thering is otherwise complete. The bezel has an oval setting for the intaglio, framed by anopenwork rim. The shoulders of the hoop are decorated with gold filigree ornament in the formof two volutes, each terminating in a bead. The volutes spring from the join between theshoulder and the rest of the hoop. Between them is a further bead. The hoop is otherwisedecorated with two parallel grooves. Rings from hoards at Silchester and Hoxne carry similarfiligree decoration.84

The intaglio, which has a flat upper surface with a bevelled surround, depicts a standing nakedadolescent with crossed legs, leaning on a short spiral column; the short wings which sprout fromhis shoulders identify him as Cupid. He leans his left arm on the column, his right being raised andholding a torch which he will later use to burn Psyche manifested as a butterfly.85 Martin Henig(pers. comm.) notes the similarity of the figure and his pose to that of Cupid portrayed on a first- orsecond-century red jasper in the Kunsthistorisches Museum in Vienna (although here Cupid leansagainst the column which is in front of him).86 He also observes similarities in the execution of theTangley gem with that of Ring 5 from the Silchester hoard and tentatively attributes both to the

FIG. 19. Eynsford, reclining Mercury figurine (No. 19). Scale 1:1. (Photo: K. Sumnall; © K. Sumnall and Museum ofLondon)

82 Durham, op. cit. (note 45), no. 35; Boucher, op. cit. (note 49), nos 403–4, pl. 81; Kaufmann-Heinimann, op. cit. (note49), 41, no. 36, Taf. 53.

83 Found by A. Duke. Identified and recorded by K. Hinds and A. Willis.84 M. Fulford, M. Henig and C. Johns, ‘A late Roman gold finger-ring from Silchester, Hampshire’, Britannia 18

(1987), 279–81; M. Fulford, A. Burnett, M. Henig and C. Johns, ‘A hoard of late Roman rings and silver coins fromSilchester, Hampshire’, Britannia 20 (1989), 222–3, ring nos 3 and 5; C. Johns, The Hoxne Late Roman Treasure,213–14, nos 8 and 9.

85 M. Henig, A Corpus of Roman Engraved Gemstones from British Sites, BAR British Series 8 (3rd edn, 2007), 105–7,nos 112–25; V. Platt, ‘Butterflies: seals, symbols and the soul in Antiquity’, in L. Gilmour (ed.), Pagans and Christians –from Antiquity to the Middle Ages. Papers in honour of Martin Henig, BAR International Series 1610 (2007), 89–99.

86 E. Zwierlein-Diehl, Die antiken Gemmen des Kunsthistorischen Museums in Wien II (1979), 182, no. 1357, Taf. 127.


so-called ‘style calligraphique’ characteristic of some fourth-century examples, for example a verylarge pale nicolo in Paris depicting Dea Roma dated to c. A.D. 330.87

DORSET

(21) Sutton Waldron (DOR-695A14) (FIG. 21).88 A copper-alloy casing for a padlock in the formof a human head. It is 31.4 mm long, 19.2 mm wide and 15.7 mm deep, and weighs 14.3 g. Thefront of the padlock takes the form of a crudely rendered human face or mask, probably female.Hair swept backwards is rendered by short incisions from the arched hairline. Thedisproportionately large lentoid eyes are thickly outlined on either side of a triangular nose. Themouth is a short horizontal slit. The keyhole is on the left side of the face and the right side ispierced by a small circular aperture below the hairline. There is a rectangular opening at the topof the head which is framed on three sides by a ridged moulding. In this opening the hinge tojoin the missing backplate may have sat. At the base of the casing is a small flat projectionwith a curved lower edge, which projects towards the front of the lock. The lock is almostidentical to examples documented by the PAS in North Yorkshire and Lincolnshire.89 A smallnumber of further casings very similar to the PAS finds are also documented in privatecollections.90 The close resemblance between the Sutton Waldron casing and similar examplessuggests that they are the likely product of a single workshop. Being intended to house thewhole of the lock mechanism, they are much deeper than other more commonly documentedfigural locks, on which a stylised and often elaborate mask forms the front plate of the padlock.91

FIG. 20. Tangley, ring and intaglio (No. 20). Scale: 1:1. (Photo: K. Hinds; © K. Hinds and Hampshire Cultural Trust)

87 E. Zwierlein-Diehl, ‘Constantinopolis et Roma. Intailles du IVe et du Ve siècle après Jesus-Christ’, in M.A. Broustet(ed.), La Glyptique des mondes classiques. Mélanges en hommage à Marie-Louise Vollenweider (1997), 92, fig. 15.

88 Found by D. Cobb. Identified and recorded by C. Hayward Trevarthen.89 YORYM-89CD53 and LIN-3BC5E5.90 We are indebted to Jerry Slocum for his observations on this category of locks. A very similar casing is documented

by him for inclusion in a forthcoming catalogue of his collection. J. Slocum and D. Sonneveld, Roman Empire Mask PuzzlePadlocks, Slocum Inv. # 34021 (formerly Donald Jackson collection 5067). He also draws our attention to a small numberof other examples in private European collections (J. Slocum, pers. comm.).

91 Feugère, op. cit. (note 19), CDN-4003. A PAS object from Hadham, Herts. (BH-6225E0), is less likely to be aseal-box than a casing for a lock of this type.


SOMERSET

(22) Somerton (SOM-6EA483) (FIG. 22).92 Part of a copper-alloy rod-shaped handle, circular insection, likely to be from a pair of Roman uvula forceps, probably from a staphylagra, i.e. atoothed-jaw type rather than a staphylocaustes (un-toothed type). What remains of the handlefrom terminal to break is 75.8 mm long and 5.8 mm in diameter (at its widest). The terminaltakes the form of an elegant baluster finial with a circular moulding where it connects to thehandle. Another circular moulding in the form of a double ridge occurs further alongthe handle. From better preserved examples it is likely that a two to three centimetres length ofthis handle has been lost beyond the break.

The forceps, which cannot be dated more precisely than to the first to fourth centuries A.D., wereprobably used for uvulectomy (removal of the uvula at the back of the mouth) andhaemorrhoidectomy. 26 examples of forceps of this type were documented across the Empiremore than 20 years ago. Other British finds include complete examples from Caerwent andDorchester, a jaw fragment from Leicester and variants from Colchester (2) and Ancaster.93

FIG. 21. Sutton Waldron, padlock casing (No. 21). Scale 1:1. (Photo: C. Hayward Trevarthen; © Somerset CountyCouncil)

FIG. 22. Somerton, forceps handle (No. 22). Scale 1:1. (Photo: L. Burnett; © L. Burnett and Somerset County Council)

92 Found by V. Macrae. Identified by R. Jackson. Recorded by L. Burnett.93 R. Jackson, ‘Staphylagra, staphylocaustes, uvulectomy and haemorrhoidectomy: the Roman instruments and

operations’, in A. Krug (ed.), From Epidaurus to Salerno: Symposium held at the European University Centre for


PAS finds comprise an increasingly important component of Roman period objects with a surgical(and toilet) function in Britain, although they have yet to be synthesised. Examples include otherforceps, for example a pointed jaw forceps of which the findspot (SOMDOR-D963B7) is not farfrom the Somerton find, and possible scalpel handles, for example octagonal-facetted handles fromScarning and Burgh Castle, Norfolk (NMS-8801B7; NMS-4157A2).94

CORRIGENDUM

A lead-alloy plaque from Clare, Suffolk (SF-301791) carrying a relief image of a statue within anarchitectural setting was interpreted as a possible representation of a shrine on a votive plaque,although no specific ancient parallels could be noted.95 The scene, however, is almost identicalto a neo-Classical architectural scene on firemarks of the early nineteenth century and the Clarepiece is clearly a fragment from a plaque of this type.96

Portable Antiquities Scheme, Institute of Archaeology, University College London (S.W.)[email protected] of Classics, King’s College London (J.P.)[email protected]

This paper is published with the aid of a grant from the Portable Antiquities Scheme

doi:10.1017/S0068113X15000446

Cultural Heritage, Ravello, April, 1990 (1992), 167–85; C. Sparey-Green, ‘A Roman medical instrument from Dorchester’,Proceedings Dorset Natural History and Archaeology Society 116 (1994), 135–6. Further references to discussion of thisobject may be found in the online record.

94 Worrell, op. cit. (note 1, 2007), 34–5, no. 30.95 Worrell and Pearce, op. cit. (note 4, 2014), 421–2, no. 23.96 We thank Laura Burnett for drawing this to our attention. The specific parallels are with early nineteenth-century

Royal Assurance Society firemarks, of which an example (with a very clear online image) was sold by Bonhams in2007: http://www.bonhams.com/auctions/14987/lot/131/.


mailto:[email protected]

mailto:[email protected]

http://www.bonhams.com/auctions/14987/lot/131/

http://www.bonhams.com/auctions/14987/lot/131/

britannia //finds.org.uk/documents/britannia/2015 (discussing 2014).pdf · britannia / volume 46 /...

Documents